Kinesin Moves by an Asymmetric Hand-Over-Hand Mechanism

Introduction
This review discusses the motion of kinesin, a double headed motor protein. A study was conducted to determine which of two motion patterns is the one which describes the movement of this protein: the inchworm model, or the hand-over-hand model.
What is Kinesin?
Kinesin is a protein in a class of motor proteins which are powered by the hydrolysis of ATP – the molecule responsible for transporting chemical energy for metabolism [1]. Kinesin transports large cargo about cells by walking along microtubules, hydrolysing one molecule of ATP per step [2]. It has been proposed than the force of the protein binding to the microtubule propels the cargo along [3]. Kinesin moves to the “plus” end of the microtubule, meaning it transports the cargo from the centre to the edge of the cell [4]. There is evidence that some kinesins have a role in mitosis (cell division), by separating microtubules or depolymerising them [5].
The Models
The inchworm model describes motion with one “arm” of the protein moving forward, followed by the other, with the first arm always in the lead. There are two types of inchworm motion, symmetric and asymmetric, which are shown in the image below. The symmetric model takes smaller steps, so only one arm moves at a time. Asymmetric motion takes a single step, at the middle of which both arms move.
In the hand-over-hand model, alternating arms move forward over each other. In the symmetric case, the molecule rotates in the same direction every time, but in the asymmetric case the molecule rotates in alternating directions. These models are shown in the image below.

Main Results
The paper’s main result shows that the kinesin protein moves using an asymmetric hand-over-hand mechanism. To reach this verdict, a variety of single molecule experiments were performed. They established that the individual kinesin dimers make discrete steps at random intervals along the microtubule, and may take as many as one hundred 8 nm steps before releasing. The movement is processive, meaning that the protein can make many consecutive steps without releasing the substrate (the molecule on which it acts – here, the microtubule). This motion exists even when external forces up to several pN are applied, which indicates part of the protein stays attached at all times.

Get Help With Your Essay
If you need assistance with writing your essay, our professional essay writing service is here to help!
Essay Writing Service

The active part of kinesin is composed of a dimer, with two identical heavy chains, each with a “head” attached to a common stalk. These chains join to a short “neck” composed of single polypeptide chains. The heavy chains are coiled round each other to allow the rotation necessary for the hand-over-hand model. This rotation is about the neck, but the motion of the heads turning would continue winding, summing until the heavy chains would join into a common stalk, preventing independent rotation.
A study was conducted [6] showing that no significant rotation occurs of the stalk during the stepping motion. For a symmetric model, a large rotation (about 180 degrees) was expected in the hand-over-hand models. The basis for the definition of “symmetric” here was in three dimensions: the structure of the kinesin and microtubule must be identical at the start and end of each ATP hydrolytic cycle, except for the two heads having swapped places [6]. An example of this is simply the dimer rotating half a revolution about an axis perpendicular to the microtubule each step [7], hence the prediction for a rotation of 180 degrees. However this was ruled out, and an inchworm model was proposed. In this, only one of the heads is active in hydrolysis, but the possibility of an asymmetric hand-over-hand motion remained. This would mean that the head and neck move in such a way that the overall rotation of the stalk is suppressed, instead alternating between two distinct structures [8].
How They Were Obtained
The step motion of individual native and recombinant (formed in the lab by combining genetic material from multiple sources) kinesin molecules was measured, using optical force-clamp apparatus. This technique uses light from a tightly focussed laser to trap small, polarisable particles in a potential well near the focal point [9]. It was found that the intrinsic stepping rates alternated between two different values for each step, meaning the molecules “limped”. The difference in steps implies there was an alternation in underlying molecular configurations, meaning the motion could not be fully symmetric (such as the inchworm and symmetric hand-over-hand motions should be). The discovery of the limp, along with other nano-mechanical properties, means the protein moves with an asymmetric hand-over-hand motion.
Single molecules of kinesin were attached to microscopic beads, serving as markers for position and as handles for external forces. An optical trap was then used to capture the individual beads that diffused whilst carrying the kinesin, which were placed near the microtubules. This was while kinesin bound and moved. The motion was then tracked using nanometer level precision. A feedback-controlled force lamp was used to apply a constant backwards load during the motion, in order to reduce the Brownian fluctuations and improve the spatiotemporal resolution. It also allowed for the kinesin to move further, taking more steps, in order to show statistical significance.
The Results
A derivative of Drosophila melanogaster kinesin (DmK401) was shown to have an obvious limp, with large time differences in the steps despite the stochastic nature (and ensuing variability). Statistical analysis showed significant differences in the average step times for both slow and fast steps. The durations of the steps were then calculated as τslow = 136 ± 6 ms and τfast = 24 ± 1 ms. The limp factor, L, can then be calculated as the ratio of the mean duration of the slow stepping time to the mean duration of the fast stepping time. The distribution showed significant limping for the majority of molecules, but there was wide variation in the results. 63% of records showed L > 4, and the average was L = 6.45 ± 0.31. Some motors took many runs and had consistently higher limp factors than others, but the distribution was broad and the populations could not be separated of limping and non-limping molecules.
Other kinesin molecules, such as the native squid kinesin, showed almost no evidence of limping – the same calculations were applied as to DmK401, and the times were calculated to be τslow = 90 ± 4 ms and τfast = 54 ± 2 ms. The difference is much smaller than that for DmK401. The limp distribution was also found to be narrower, with the average limp factor being L = 2.23 ± 0.14, only slightly higher than the estimated value for a non-limping molecule, L ~ 1.8.
The test was then done with kinesin derivates of Drosophila which had increasing stalk lengths. Longer stalks mean the motors are less likely to limp. The largest stalk tested was that of DmK871, and this had a limp factor of L = 2.16 ± 0.17, which was indistinguishable from native squid kinesin. There was also a correlation between an increasing limp factor (therefore shorter stalks) and an increase in characteristic lifetime of the slow step time, whereas the fast step remained invariant. This suggests the limping comes from one head alone, and the other is indifferent.
A bacterial expression of a derivative of human kinesin (HsK413) also limped, with limp factor = 2.98 ± 0.25, much greater than the native squid kinesin, but still less than DmK401 and DmK448. Rarely, squid kinesin molecules seemed to limp, making outliers – some of which limped consistently.
Discussion
As both native and bacterially expressed dimers from different species can limp, this behaviour may be a result of a common mechanism describing how all kinesin molecules move. The alternation between short and long step times during limping reflects an alternation between the intrinsic rate (the rate with which the population increases) and the time it takes to leave each phase where neither head is moving. This implies the structure of the kinesin-microtubule complex is different at the end of sequential steps. The mechanism describing the movement of kinesin must therefore be asymmetric, meaning the molecular configuration switches after each step. Symmetric mechanisms, by definition, cannot account for switching – inchworm models will not limp without additional (asymmetric) features, nor will symmetric hand-over-hand models.
The detail of how kinesin motors move is not well known or understood, so we cannot look at how limping could relate to the structure of the motion, but there are some suggestions based on the asymmetric hand-over-hand mechanism. Limping could be caused by misalignment of the stalk coils, meaning the necks would be different lengths, hence the head with a shorter neck would need extra time to find the next binding site using a diffusional search and overall slowing the kinetics. Another option is that there could be over- or under-winding of the coils from hand-over-hand motion, causing torsional asymmetry. The energy required to coil or uncoil the stalk would be reduced, changing the equilibrium and the rate with which the head moves forward.
Whilst there is no immediate explanation for the effect whereby the shorter stalks result in longer slow stepping times, it may be incorporated into later studies with further assumptions. However, these experiments have shown that more approaches are needed for single-molecule experiments to answer these questions.
Despite the exact mechanism not being known, the experiments do show that the kinesin motors limp, and making the asymmetric hand-over-hand mechanism the most likely.
Why is this Significant?
This is a breakthrough in the field, as more detail can now be found on how biological motors move. By establishing how kinesin moves, other motors can be analysed to find their mechanisms for movement, and this helps to further our understanding of biology. The same experiment can be done with other proteins, or more experiments can be done with kinesin to better understand certain factors – for example, why the slow stepping times correlate to shorter stalks, or why limping appears to come from one head only.
What Other Work has Been Done?
Whilst many similar experiments have been conducted, this has improved the knowledge of kinesin’s movement by showing that the inchworm model does not apply. It has opened up avenues for either looking deeper into kinesin’s movement, or for finding the mechanisms for other motors. The experimental method used was also new, and was different from the previous study by Hua, Chung and Gelles [6], which was researching something similar.
Hua, Chung and Gelles looked at the rotations in the movement of kinesin, with the null hypothesis of a symmetric hand-over-hand mechanism. This was done by immobilizing a derivative of Drosophila, and measuring the different orientations of the microtubules as it moved. Their findings were consistent with the inchworm model, which is why it was proposed initially.
Conclusion
The kinesin mechanism is now better understood than before, with more information gained in explaining the mechanism. This has presented more factors for consideration and created new questions to answer: the opportunity for further research is huge. Future experiments could consider different protein’s movements, or look deeper into the kinesin mechanism. The discovery is significant as it has not only given more insight into biological motors, but revealed ample possibilities for more experiments in the field.
Acknowledgement
This review was written primarily based on the work of “Asbury, C L et al. 2003. Kinesin Moes by an Asymmetric Hand-Over-Hand Mechanism. Science. 302(2130).”, with all numerical values and the majority of content based on the paper, unless otherwise stated.
References
[1] Knowles, JR. 1980. Enzyme-catalyzed phosphoryl transfer reactions. Annual Revue of Biochemistry. 49, pp.877–919.
[2] Schnitzer MJ. 1997. Kinesin hydrolyses one ATP per 8-nm step. Nature. 388(6640), pp.386–390.
[3] Mather, WH and Fox RF. 2006. Kinesin’s biased stepping mechanism: amplification of neck linker zippering. Biophysical Journal. 91(7), pp.2416–26.
[4] Ambrose, JC, et al. 2005. A minus-end-directed kinesin with plus-end tracking protein activity is involved in spindle morphogenesis. Molecular Biology of the Cell . 16(4) pp.1584–92.
[5] Goshima, G and Vale, RD. 2005. Cell cycle-dependent dynamics and regulation of mitotic kinesins in Drosophila S2 cells. Molecular Biology of the Cell. 16(8) pp.3896–907.
[6] Hua, W, Chung J, and Gelles, J. 2002. Science. 295, p.844.
[7] Howard, J. 1996. Annual Revue of Physiology. 58, p.703.
[8] Hoenger, A et al. 2000. Journal of Molecular Biology. 297, p.1087.
[9] Greenleaf, W J et al 2005. Physical Revue Letters. 95, 208102.